Short answer: Lead scoring is a simple way to rank leads by how likely they are to buy, so your team calls the best ones first. For a small team, pick four or five signals you already capture, such as fit, source and recent replies, give each a few points, add them up, and review the scores against real wins every month.

Most small teams don’t need a scoring engine. They need a shared answer to one question: which lead do we call next? Lead scoring gives you that answer in a form everyone can see and argue with. This guide walks through a plain points model you can run on a whiteboard, a spreadsheet or a CRM, how to pick the signals, and how to tell whether the scores are working. If you’re still getting leads into one place, start with our guide to building a lead tracking system, then come back here.

Where I’m coming from: I build CRM software, including EloSync, and lead ranking is one of the first things people ask about when they move off a spreadsheet. The model below is the one I’d set up for a team of two to ten people. It’s deliberately boring. Boring models get used.

What is lead scoring?

Lead scoring is giving each lead a number based on who they are and what they’ve done, so the list can be sorted by priority instead of by date.

The number itself means nothing outside your team. A score of 60 isn’t “good” in any absolute sense. It’s only useful because a lead at 60 should, on average, be a better use of the next call than a lead at 20. That’s the whole idea. Everything else, the points, the categories, the thresholds, is there to make that ranking honest and repeatable.

Scores usually come from two kinds of information:

• Fit: does this lead look like the people who buy from you? Think company type, size, role, budget range or location of service, whatever actually matters in your sales.
• Interest: are they acting like someone who wants to buy? Think replying to a message, asking about pricing, booking a call or filling in a detailed form.

A lead with strong fit and strong interest goes to the top. A lead with neither drops to the bottom, and you stop feeling guilty about not calling them today.

Do small teams actually need lead scoring?

Yes, once you have more leads than you can properly follow up in a day. Below that point, a short list sorted by date is fine.

The trouble usually starts quietly. A campaign brings in a burst of leads, someone works the list from the top, and the best lead of the week sits at row forty for three days. Speed matters a lot here. We cover the reasons in our post on lead response time, but the short version is that interest fades while a lead waits. Scoring doesn’t make you faster. It makes sure your speed goes to the right people.

There’s a second, less obvious benefit. Writing down a scoring model forces the team to agree on what a good lead looks like. In my experience that conversation is often more useful than the scores themselves, because it surfaces the quiet disagreements: the salesperson who loves small accounts, the founder who only wants big ones.

How do you build a simple lead scoring model?

Pick a handful of signals you already capture, give each one a small number of points, set two or three score bands, and test the result against leads you’ve already won or lost.

Here’s the process step by step.

1. Look at your last batch of won deals

Before you assign a single point, pull up the last twenty or so customers you closed, or as many as you have. Write down what they had in common when they first arrived. Where did they come from? What kind of business were they? What did they say in their first message? Then do the same for a batch of leads that went nowhere.

You’re not doing statistics here. You’re looking for the obvious patterns. If almost every won deal came from referrals and the website contact form, and almost none came from a cold list, that’s a signal worth scoring.

2. Choose four or five signals

Keep it small. Every extra signal is one more field someone has to fill in correctly, and one more thing to argue about. Good signals for a small team are ones that are already being captured, or that take a second to record:

• Lead source, such as referral, website form, Facebook lead form or event. If you aren’t recording this yet, our guide to lead source tracking shows how.
• Fit with your ideal customer, scored as a simple yes, partly or no.
• Stated need or timeline, such as “needs this soon” versus “just looking”.
• Two-way contact, meaning they’ve replied or picked up the phone at least once.
• A buying question, such as asking about price, contract terms or a start date.

Leave out anything you can’t capture reliably. A signal that’s filled in for half your leads will quietly skew the ranking.

3. Give each signal a few points

Use small, round numbers. The exact values matter far less than the order they create. Here’s an example model to start from. The points are illustrative, not a benchmark, and you should adjust them to match your own won deals.

Signal	Condition	Example points
Lead source	Referral or direct enquiry	+20
Lead source	Website or Facebook lead form	+10
Fit	Matches your ideal customer	+25
Fit	Partly matches	+10
Timeline	Says they need it soon	+15
Two-way contact	Has replied or answered a call	+15
Buying question	Asked about price, terms or start date	+20
Silence	No reply after your full follow-up sequence	-20

Example data for illustration only.

Notice the negative row. Scores that only ever go up end up flattering old leads that went cold months ago. A small penalty for silence keeps the top of the list fresh.

4. Set two or three bands

Nobody acts on “this lead is a 47”. People act on “this lead is hot”. So turn the number into bands, for example:

• Hot: call today.
• Warm: follow up this week as part of your normal cadence.
• Cold: keep in a slower nurture list, or close it out.

Where you draw the lines depends on your points, so set them after you’ve scored a test batch, not before.

5. Score old leads before you trust it

Take the won and lost leads from step one and score them with your new model. Did the won deals mostly land in hot and warm? Did the lost ones mostly land in cold? If a lot of your best customers would have been marked cold, the model is wrong, and it’s much cheaper to find that out now than after a month of ignoring the wrong people.

Which lead scoring signals should you avoid?

Avoid signals you can’t verify, signals that are really your own activity, and signals that reward one salesperson’s habits over the lead’s behaviour.

A few traps I see often:

• Counting your own outreach. “We’ve called them five times” isn’t interest. It’s effort. If anything, lots of unanswered attempts should lower a score.
• Gut-feel fields. A “quality: 1 to 5” dropdown that each person fills in differently turns the score into a popularity contest.
• Too many tiny signals. Twelve signals worth three points each produce scores nobody can explain. If you can’t say why a lead is hot in one sentence, the model is too complicated.
• Signals you can’t capture yet. Don’t design around data you plan to collect someday. Score what you have today.

Lead scoring vs lead routing: what’s the difference?

Scoring decides how urgent a lead is. Routing decides who handles it. They work best together, but they’re separate decisions.

A common setup is to route first, by region, product or round robin, and then let each owner sort their own list by score. Another is to send hot leads straight to your most experienced closer. Either way, keep the two rules written down separately so you can change one without breaking the other. We cover the routing side in detail in our guide to lead routing.

How do you run lead scoring in a CRM?

Store each signal as a field or label on the lead, record the score where everyone can see it, and sort or filter your lead list by it.

You don’t need a dedicated scoring feature to start. In most CRMs, including EloSync, a lead record can carry the details you’re scoring on, and your team can update a score field as they qualify each lead. What matters is that the signals live on the lead itself rather than in someone’s head or a side spreadsheet.

A few practical notes for a small-team setup:

• Capture source automatically where you can. In EloSync, leads can come in from Meta Lead Ads (Facebook lead forms), through a webhook from your own website form, or be added manually. Leads that arrive through a connected form already know where they came from, which saves one field of manual entry.
• Turn the score into a task. A hot lead should create a follow-up task for today. EloSync’s Leads, Tasks and ToDos are free, and tasks sit on a Kanban board, so hot-lead calls are easy to see alongside the rest of the day’s work.
• Use roles to keep the model stable. Let everyone see scores, but limit who can change the scoring rules. With roles, you can decide who edits what.
• Calculate it elsewhere if you need to. If you want scores worked out automatically from your own data, a developer can use the REST API and webhooks to read lead details and write a score back. Most small teams won’t need this at first.

If you’re building your own system instead of buying one, my walkthrough on how to build a CRM covers the data model a scoring field would sit on.

How often should you review your lead scores?

Once a month is enough for most small teams. Each review, compare the scores of recent wins and losses and adjust one thing at a time.

Keep the review short. Pull the leads you closed and lost in the last month and ask three questions:

1. Were most wins scored hot or warm when they arrived?
2. Did any cold leads turn into customers, and what did they have in common?
3. Is any signal almost always the same for every lead? If so, it isn’t helping you tell leads apart, so drop it.

Change one weight at a time and write down what you changed and why. If you change four things in one go, you’ll never know which one helped. Your weekly numbers, such as how many hot leads got a call the same day, belong in your regular sales team KPIs review.

What does a lead scoring week look like in practice?

It should feel like a normal sales week with a better-sorted list, not a new process.

Here’s how I’d expect it to run. New leads arrive and get their source recorded on intake. Whoever qualifies a lead marks fit and timeline during the first conversation. Replies and buying questions get noted as they happen, which nudges the score up. Each morning, the team opens the lead list sorted by score, calls the hot ones, and works warm leads through the usual follow-up cadence. Leads that go silent after the full sequence lose points and drift down. Once a month, someone spends half an hour checking the model against real results.

That’s it. If your scoring takes more time than that each week, it has grown too complicated for the size of your team.

Common lead scoring mistakes

• Building the model before looking at real wins. Start from your own customers, not a template.
• Never lowering scores. Without decay or penalties, old leads crowd the top.
• Hiding the logic. If the team doesn’t understand why a lead is hot, they won’t trust the list and will go back to gut feel.
• Treating the score as the truth. A score is a sorting aid. A salesperson who hears a real buying signal on a call should act on it, whatever the number says.
• Scoring and never acting. A hot lead with no follow-up task is just a nicer-looking spreadsheet row.

If you want a simple place to keep leads, scores and follow-up tasks together, you can join the EloSync beta. Leads, Tasks and ToDos are free.

FAQ

What is a good lead score?

There’s no universal good score. A score only means something inside your own model. What matters is that leads scored higher turn into customers more often than leads scored lower, which you check by scoring past wins and losses.

How many signals should a lead scoring model have?

For a small team, four or five is plenty. Fewer signals are easier to capture consistently and easier to explain, and a model the team understands is one they’ll actually use.

Should lead scores go down over time?

Yes. Add a penalty for silence after your full follow-up sequence, or lower scores for leads with no activity in a set period. Otherwise old, cold leads stay at the top of the list.

Can I do lead scoring in a spreadsheet?

You can, and it’s a fine way to test a model. It gets harder once several people are updating the same sheet, which is usually the point where a CRM makes more sense. Our Excel vs CRM comparison covers that trade-off.

Is lead scoring the same as lead qualification?

Not quite. Qualification is deciding whether a lead is a real opportunity at all. Scoring ranks the leads you have so you know which to work first. Many teams use the score to decide which leads to qualify first.